ESNemble: an Echo State Network-based ensemble for workload prediction and resource allocation of Web applications in the cloud

2019 ◽  
Vol 75 (10) ◽  
pp. 6303-6323 ◽  
Author(s):  
Hoang Minh Nguyen ◽  
Gaurav Kalra ◽  
Tae Joon Jun ◽  
Sungpil Woo ◽  
Daeyoung Kim
Keyword(s):  
Resource Allocation ◽  
Web Applications ◽  
Echo State Network ◽  
Workload Prediction

scholarly journals Exploratory Learning of Resource Management in Private Cloud Environment

2019 ◽  
Vol 8 (3) ◽  
pp. 8011-8014
Keyword(s):  
Cloud Computing ◽  
Web Applications ◽  
Service Level Agreement ◽  
Service Level ◽  
Private Cloud ◽  
Workload Prediction ◽  
Machine Learning Methods ◽  
Optimized Model ◽  
Auto Scaling ◽  
Near Future

Paper: Scientific and Web applications are major sources of Internet traffic that requires resources such as Memory ,CPU and Network are on demand. Cloud computing and virtualization are the boons for such resource demand applications from various users. Service models of cloud computing provide a platform for many applications to use resources as pay per use model. In Cloud, Auto-scaling with manage Service Level Agreement (SLA) of resources is one of the main challenges to meet the current demand for resources. To maintain the performance of the cloud, which provision resources based on a heuristic for workload prediction is prime importance. In this paper, we address auto-scaling as a problem to forecast near-future demand of resource using a KNN machine learning methods suggest the optimized model for the dynamic variation of CPU utilization

A new method for human resource allocation in cloud-based e-commerce using a meta-heuristic algorithm

Kybernetes ◽  
2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ibrahim Al-Shourbaji ◽  
Waleed Zogaan
Keyword(s):  
Resource Allocation ◽  
Human Resource ◽  
Polynomial Time ◽  
Web Applications ◽  
Performance Enhancement ◽  
Imperialist Competitive Algorithm ◽  
Allocation Problem ◽  
Content Type ◽  
Capital Costs ◽  
Competitive Algorithm

PurposeThe human resource (HR) allocation problem is one of the critical dimensions of the project management process. Due to this nature of the problem, researchers are continually optimizing one or more critical scheduling and allocation challenges in different ways. This study aims to optimize two goals, increasing customer satisfaction and reducing costs using the imperialist competitive algorithm.Design/methodology/approachCloud-based e-commerce applications are preferred to conventional systems because they can save money in many areas, including resource use, running expenses, capital costs, maintenance and operation costs. In web applications, its core functionality of performance enhancement and automated device recovery is important. HR knowledge, expertise and competencies are becoming increasingly valuable carriers for organizational competitive advantage. As a result, HR management is becoming more relevant, as it seeks to channel all of the workers’ energy into meeting the organizational strategic objectives. The allocation of resources to maximize benefit or minimize cost is known as the resource allocation problem. Since discovering solutions in polynomial time is complicated, HR allocation in cloud-based e-commerce is an Nondeterministic Polynomial time (NP)-hard problem. In this paper, to promote the respective strengths and minimize the weaknesses, the imperialist competitive algorithm is suggested to solve these issues. The imperialist competitive algorithm is tested by comparing it to the literature’s novel algorithms using a simulation.FindingsEmpirical outcomes have illustrated that the suggested hybrid method achieves higher performance in discovering the appropriate HR allocation than some modern techniques.Practical implicationsThe paper presents a useful method for improving HR allocation methods. The MATLAB-based simulation results have indicated that costs and waiting time have been improved compared to other algorithms, which cause the high application of this method in practical projects.Originality/valueThe main novelty of this paper is using an imperialist competitive algorithm for finding the best solution to the HR allocation problem in cloud-based e-commerce.

Auto-Scaling Provision Basing on Workload Prediction in the Virtualized Data Center

2020 ◽  
Vol 12 (1) ◽  
pp. 53-69 ◽  
Author(s):  
Danqing Feng ◽  
Zhibo Wu ◽  
Decheng Zuo ◽  
Zhan Zhang
Keyword(s):  
Resource Allocation ◽  
Service Level Agreement ◽  
Minimum Cost ◽  
Pso Algorithm ◽  
Service Level ◽  
Workload Prediction ◽  
Cloud Data ◽  
Combination Forecast ◽  
Application Service ◽  
Forecast Technique

With the development in the Cloud datacenters, the purpose of the efficient resource allocation is to meet the demand of the users instantly with the minimum rent cost. Thus, the elastic resource allocation strategy is usually combined with the prediction technology. This article proposes a novel predict method combination forecast technique, including both exponential smoothing (ES) and auto-regressive and polynomial fitting (PF) model. The aim of combination prediction is to achieve an efficient forecast technique according to the periodic and random feature of the workload and meet the application service level agreement (SLA) with the minimum cost. Moreover, the ES prediction with PSO algorithm gives a fine-grained scaling up and down the resources combining the heuristic algorithm in the future. APWP would solve the periodical or hybrid fluctuation of the workload in the cloud data centers. Finally, experiments improve that the combined prediction model meets the SLA with the better precision accuracy with the minimum renting cost.

TASM: technocrat ARIMA and SVR model for workload prediction of web applications in cloud

2018 ◽  
Vol 22 (2) ◽  
pp. 619-633 ◽  
Author(s):  
Parminder Singh ◽  
Pooja Gupta ◽  
Kiran Jyoti
Keyword(s):  
Web Applications ◽  
Workload Prediction
Sign in / Sign up

Export Citation Format

Share Document